A Stochastic Gradient Approach for Robust Heartbeat Detection with Doppler Radar Using Time-Window-Variation Technique

Chen Ye, Kentaro Toyoda, Tomoaki Ohtsuki

Research output: Contribution to journalArticle

Abstract

Heart rate variability (HRV) indicates health condition and mental stress. The development of non-contact heart rate (HR) monitoring technique with Doppler radar is attracting great attentions. However, the performance of heartbeat detection via radar signal easily degrades, due to respiration and body motion. In this paper, first, a stochastic gradient approach is applied to reconstruct high-resolution spectrum of heartbeat, by proposing the zero-attracting sign least-mean-square (ZA-SLMS) algorithm. To correct the quantized gradient of cost function, and penalize the sparse constraint on the updating spectrum, more accurate heartbeat spectrum is reconstructed. Then, to better adapt to the noises with different strengths caused by subjects' movements, an adaptive regularization parameter (AREPA) is introduced in the ZA-SLMS algorithm as an improved variant, which can adaptively regulate the proportion between gradient correction and sparse penalty. Moreover, in view of the stability of location of spectral peak associated with HR when the size of time window slightly changes, a time-window-variation (TWV) technique is further incorporated in the improved ZA-SLMS (IZA-SLMS) algorithm, for more stable HR estimation. Through the experiments on five subjects, our proposal is demonstrated to bring a significant improvement of accuracy against existing detection methods. Specifically, the IZA-SLMS algorithm with TWV achieves the smallest average error of 3.79 beats per minute (BPM), when subjects type with a laptop.

Original languageEnglish
JournalIEEE Transactions on Biomedical Engineering
DOIs
Publication statusAccepted/In press - 2018 Jan 1

Fingerprint

Doppler radar
Cost functions
Radar
Health
Monitoring
Experiments

Keywords

  • adaptive filter
  • Continuous wavelet transforms
  • Doppler radar
  • Doppler radar
  • Heart beat
  • heart rate (HR)
  • Heart rate variability
  • Monitoring
  • Non-contact heartbeat detection
  • sparse spectrum reconstruction (SSR)
  • time-window-variation (TWV)

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

@article{ca826c7ff16e4660b15f2118f2de8d98,
title = "A Stochastic Gradient Approach for Robust Heartbeat Detection with Doppler Radar Using Time-Window-Variation Technique",
abstract = "Heart rate variability (HRV) indicates health condition and mental stress. The development of non-contact heart rate (HR) monitoring technique with Doppler radar is attracting great attentions. However, the performance of heartbeat detection via radar signal easily degrades, due to respiration and body motion. In this paper, first, a stochastic gradient approach is applied to reconstruct high-resolution spectrum of heartbeat, by proposing the zero-attracting sign least-mean-square (ZA-SLMS) algorithm. To correct the quantized gradient of cost function, and penalize the sparse constraint on the updating spectrum, more accurate heartbeat spectrum is reconstructed. Then, to better adapt to the noises with different strengths caused by subjects' movements, an adaptive regularization parameter (AREPA) is introduced in the ZA-SLMS algorithm as an improved variant, which can adaptively regulate the proportion between gradient correction and sparse penalty. Moreover, in view of the stability of location of spectral peak associated with HR when the size of time window slightly changes, a time-window-variation (TWV) technique is further incorporated in the improved ZA-SLMS (IZA-SLMS) algorithm, for more stable HR estimation. Through the experiments on five subjects, our proposal is demonstrated to bring a significant improvement of accuracy against existing detection methods. Specifically, the IZA-SLMS algorithm with TWV achieves the smallest average error of 3.79 beats per minute (BPM), when subjects type with a laptop.",
keywords = "adaptive filter, Continuous wavelet transforms, Doppler radar, Doppler radar, Heart beat, heart rate (HR), Heart rate variability, Monitoring, Non-contact heartbeat detection, sparse spectrum reconstruction (SSR), time-window-variation (TWV)",
author = "Chen Ye and Kentaro Toyoda and Tomoaki Ohtsuki",
year = "2018",
month = "1",
day = "1",
doi = "10.1109/TBME.2018.2878881",
language = "English",
journal = "IEEE Transactions on Biomedical Engineering",
issn = "0018-9294",
publisher = "IEEE Computer Society",

}

TY - JOUR

T1 - A Stochastic Gradient Approach for Robust Heartbeat Detection with Doppler Radar Using Time-Window-Variation Technique

AU - Ye, Chen

AU - Toyoda, Kentaro

AU - Ohtsuki, Tomoaki

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Heart rate variability (HRV) indicates health condition and mental stress. The development of non-contact heart rate (HR) monitoring technique with Doppler radar is attracting great attentions. However, the performance of heartbeat detection via radar signal easily degrades, due to respiration and body motion. In this paper, first, a stochastic gradient approach is applied to reconstruct high-resolution spectrum of heartbeat, by proposing the zero-attracting sign least-mean-square (ZA-SLMS) algorithm. To correct the quantized gradient of cost function, and penalize the sparse constraint on the updating spectrum, more accurate heartbeat spectrum is reconstructed. Then, to better adapt to the noises with different strengths caused by subjects' movements, an adaptive regularization parameter (AREPA) is introduced in the ZA-SLMS algorithm as an improved variant, which can adaptively regulate the proportion between gradient correction and sparse penalty. Moreover, in view of the stability of location of spectral peak associated with HR when the size of time window slightly changes, a time-window-variation (TWV) technique is further incorporated in the improved ZA-SLMS (IZA-SLMS) algorithm, for more stable HR estimation. Through the experiments on five subjects, our proposal is demonstrated to bring a significant improvement of accuracy against existing detection methods. Specifically, the IZA-SLMS algorithm with TWV achieves the smallest average error of 3.79 beats per minute (BPM), when subjects type with a laptop.

AB - Heart rate variability (HRV) indicates health condition and mental stress. The development of non-contact heart rate (HR) monitoring technique with Doppler radar is attracting great attentions. However, the performance of heartbeat detection via radar signal easily degrades, due to respiration and body motion. In this paper, first, a stochastic gradient approach is applied to reconstruct high-resolution spectrum of heartbeat, by proposing the zero-attracting sign least-mean-square (ZA-SLMS) algorithm. To correct the quantized gradient of cost function, and penalize the sparse constraint on the updating spectrum, more accurate heartbeat spectrum is reconstructed. Then, to better adapt to the noises with different strengths caused by subjects' movements, an adaptive regularization parameter (AREPA) is introduced in the ZA-SLMS algorithm as an improved variant, which can adaptively regulate the proportion between gradient correction and sparse penalty. Moreover, in view of the stability of location of spectral peak associated with HR when the size of time window slightly changes, a time-window-variation (TWV) technique is further incorporated in the improved ZA-SLMS (IZA-SLMS) algorithm, for more stable HR estimation. Through the experiments on five subjects, our proposal is demonstrated to bring a significant improvement of accuracy against existing detection methods. Specifically, the IZA-SLMS algorithm with TWV achieves the smallest average error of 3.79 beats per minute (BPM), when subjects type with a laptop.

KW - adaptive filter

KW - Continuous wavelet transforms

KW - Doppler radar

KW - Doppler radar

KW - Heart beat

KW - heart rate (HR)

KW - Heart rate variability

KW - Monitoring

KW - Non-contact heartbeat detection

KW - sparse spectrum reconstruction (SSR)

KW - time-window-variation (TWV)

UR - http://www.scopus.com/inward/record.url?scp=85055874783&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85055874783&partnerID=8YFLogxK

U2 - 10.1109/TBME.2018.2878881

DO - 10.1109/TBME.2018.2878881

M3 - Article

JO - IEEE Transactions on Biomedical Engineering

JF - IEEE Transactions on Biomedical Engineering

SN - 0018-9294

ER -